Chapter 2 Critical Phenomena and Fluid Theories

The critical state provides a valuable insight into the general phase behaviour of a

fluid and is closely linked with the nature and strength of intermolecular interactions.

In this chapter, we discuss the pure component critical point, critical lines of binary

mixtures, and the criteria for the critical state for multicomponent mixtures. We also

briefly address both classical and non-classical concepts of critical behaviour.

Calculation procedures and theor y of fluids will be given.

2.1 Critical Point of Pure Components

The theory of critical phenomena was developed from the nineteenth century, and it

gradually was understood clearly in the middle of the twentieth century. Within the

Supercritical
P fluid region
L-S
C
Pc Liquid
Solid Phase Phase
V-L
Pr
Tr
Vapour
V-S
Phase

Tr Tc T
Figure 2.1 Pure component phase diagram

thermodynamic approach, a phase transition is considered as an abrupt change of one

phase into another by changing parameters of the state such as temperature, volume

and pressure. Usually a phase can have three thermodynamic statesstable,

metastable and unstable. In phase equilibrium, for a given set of fixed parameters

there always exists a thermodynamic potential having a minimum in the states. There

14
is an important point which is called critical point, at which the coexisting phases

become indistinguishable. Figure 2.1 illustrates the phase diagram of a pure

component in the p-T projection. Vapour-solid, vapour-liquid and liquid-solid two

phase lines separate the diagram into three fieldsvapour, liquid and solid phases.

The point that is met by three lines is the triple point. The point c is the critical point,

Tc, Vc and pc respectively present critical temperature, volume and pressure, and the

area of T > Tc is the supercritical fluid region. At the critical point of a pure fluid,

p 2 p
=0
= (2.1)
V T V T
2

This conclusion is very useful in that it allows the evaluation of two parameters of an

equation of state in term of critical properties. In contrast, the critical conditions for

binary and multicomponent mixtures are much more complicated.

2.2 Critical Phenomena of Binary Mixtures

For binary and other multicomponent fluids like in the case of pure substances, the

critical states are identified as the condition at which the properties of different

coexisting phases become indistinguishable. The existence of a second component

can generate equilibria between different liquid phases in addition to vapour -liquid

equilibria. The addition of a second component also generates a variety of critical

phenomena. Figure 2.2, illustrates the simplest phase diagram of binary mixtures. The

line 1 ending at c1 is the vapour-liquid coexistence curve of pure component 1, and

the line 2 ending at c2 is vapour-liquid coexistence curve of pure component 2. The

lines with Vx1x2 and Lx1x2 are the bubble point and dew point of a mixture with mole

15
fraction of component 2, x2. x1 is the mole fraction of component 1, x1 = 1- x2 . The

points c1 , c2 and cx1x2 indicate critical points of pure component 1, pure component 2

and a mixture with mole fraction x1 and x2 . x1 + x2 = 1, and x1 , x2 [0,1],

respectively. If x2 is equal to 0, cx1x2 is c1 that is critical point of component 1. If x2

equal to 1, cx1x2 is c2 that is critical point of component 2. While x2 changes in the

range [0,1], the critical points cx1x2 form a binary critical line.

p Type I
C1

Cx1x2
Lx1x2

C2
V+L
1 Vx1x2

Figure 2.2. Pressure-temperature projection showing

vapour-liquid critical line of a simple binary mixture
binary mixture (type I).

The most important illustration of the role of the critical state is the phase behaviour

classification proposed by van Konynenburg and Scott (1980). There are six main

types of critical curves in the pressure-temperature projection (Figure 2.3). These six

main types of critical diagrams are listed as following:

Type I. A vapour -liquid critical line connects the critical points of the two pure

components.

16
Type II. Critical curve of type I with an additional liquid-liquid critical line, which

start from upper critical end point (UCEP) and extends rapidly to high pressures.

Type III. There is a vapour -liquid critical line starting from one pure component

critical point which extends to an UCEP at the end of a three phase liquid-liquid-

vapour line. Another vapour-liquid critical line starts from the critical point of the

other pure component connecting with a liquid-liquid critical line to very high

pressures.

Type IV. There are two distinct vapour -liquid critical lines starting from either pure

component critical point. A three-phase vapour -liquid-liquid line connects the other

end of the lines, the two end points respectively represent an UCEP and a lower

critical end point (LCEP). There is also a liquid-liquid critical line at low

temperatures, which ends at another UCEP.

Type V. The critical curve is similar to type IV but without the low temperature

liquid-liquid critical line.

Type VI. Critical curve of type I with a liquid-liquid critical line that is observed to

rise and fall between a LCEP and an UCEP located at either end of a three-phase line.

Recently, some new type of critical curves - closed loop phase behaviour (van Pelt et

al., 1991; Yelash and Kraska, 1998; Wang et al., 2000) have enriched the

classification of phase behaviour (Figure 2.3),

17
Type VII. This phase behaviour has the same liquid-liquid immiscibility critical line

as type VI phase behaviour, but the liquid-vapour critical line is composed of two

sections, ending and UCEP and LCEP at either end of a three-phase liquid-liquid-

vapour line.

Type VIII. Critical curve of type III with an additional liquid-liquid critical line,

which starts from lower critical end point (LCEP) to high pressures (van Pelt et al.,

1991).

Type Vm. The critical line is interrupted by a three-phase line, and exhibits a pressure

maximum and a pressure minimum.

In some cases (eg., type I and type II), sub-classes critical curves depicted in figure

2.3 can be identified. These sub-classes depend on the positions of the critical locus.

The characteristics of the phase behaviour classification are discussed in greater detail

in Chapter 4.

18
 I II III

C
C C C
C UCEP
1 C
1
P 1 LLV
2
UCEP 2
2
IV V VI

C UCEP C UCEP
LLV C LLV
C
C UCEP C
1 1
LCEP LCEP 1
UCEP LLV
2 2 LCEP 2
LLV

VII Vm VIII

C UCEP C UCEP
C C UCEP C
C
LCEP 1 LLV
1 UCEP LLV
2
2 LCEP
LCEP 2
LCEP

T
Figure 2.3 Phase behaviour classification scheme for binary mixtures. Type I, II, III, IV, V, VI and VIII
behaviour is observed experimentally whereas behaviour of Type Vm and VII has only been reported
in the literature from calculations. Critical equilibria of the binary mixtures (), the critical points of
the pure components (C), the vapour pressure curves (1, 2) and the three-phase liquid-liquid-vapour (- -
- - ) equilibria are illustrated.

2.3 Criteria for the Critical State

The classical critical conditions of an m-component fluid mixture are due to Gibbs

(Gibbs, 1928; Bumstead and van Name, 1906; Donnan and Hass, 1936). A critical

transition is observed when the boundary between stable/metastable states (spinodal

19
curve) and the coexistence boundary between different stable phases (binodal curve)

meet (Sadus, 1994). The critical transition can be geometrically interpreted as a point

of inflection of the chemical potential with respect to composition. Therefore, in

critical curves, the first and second derivatives of chemical potential with respect to

composition equal zero, and a higher derivative respect to composition must be

positive to obtain a stable critical transition. Gibbs (1928) showed (Donnan and Hass,

1936) that any thermodynamic function could be transformed from these critical

criteria. Other workers (Beegle et al., 1974; Heidemann, 1975; Reid and Beegle,

1977; Modell, 1977; Sadus, 1992) have extensively described the nature of this

transformation.

The critical properties of a m-component fluid can be obtained by determining the

temperature, volume and composition which satisfy the following conditions,

2A 2A 2A

2
...
V T x1V T xm 1V T
2A 2 A 2A
2 ...
x1V T x1 T ,V x1x m1 T ,V
W = . . . =0 (2.2)

. . .
2A 2A 2 A
... 2

m1 T
x V
m1 1 T ,V
x x xm 1 T ,V

20
 W W W
...
V T x 1 T ,V x m 1 T ,V
2A 2 A 2A

2
...
x1 V T x1 T ,V x 1 X m 1 T ,V
X = . . . =0 (2.3)

. . .
2A 2A 2A
...
x m 1 V x1 X m 1 x m 1
2
T T ,V T ,V

and

X X 2X
...
V T x 1 T ,V x m 1 T ,V
2A 2A 2A
2 ...
x 1 V x1 T ,V x 1 x m 1 T ,V
Y = . . . >0 (2.4)

. . .
2A 2A 2 A
... 2
x m 1 V T x1 x 2 T ,V x m 1 T ,V

where A, T, V and x denote the molar Helmholtz function, temperature, molar volume

and mole fraction, respectively. Equations 2.2 to 2.4 represent the simplest form of

the classical critical criteria in terms of the Helmholtz function, but they are not

unique, Sadus (1992a) and Modell (1977) discuss this aspect in greater detail.

When m = 1, equations 2.2 to 2.4 are well known as conditions for a critical point of a

one component fluid, and can be simplified to equation 2.1. If m equals 2, the

equations 2.2 to 2.4 can be simplified to:

21
 2A 2A

2

V T x1V T
W = =0 (2.5)
2A 2 A
2
x
x1V T 1 T ,V

W W

V T x1 T ,V
X = =0 (2.6)
2A 2A
2
x V x
1 T 1 T ,V

and

X X

V T x1 T ,V
Y = >0 (2.7)
2A 2A
2
x V x
1 1 T ,V

Spear et al. (1969), Peng and Robinson (1977), Hicks and Young (1977), Sadus et al.

(1988) and Wang et al. (2000) calculated critical curves of binary mixtures using

equations 2.5 to 2.7. There are relatively few examples of ternary and other

multicomponent critical point calculations (Peng and Robinson, 1977; Spear et al.,

1971; Sadus and Young, 1987, 1988; Wisniak, 1983, 1984a, b; Sadus, 1992a, b and

Wei, 1998).

Calculations for binary mixtures are commonly reported to satisfy equations 2.5 and

2.6, but condition 2.7 is less frequently applied (Sadus 1994). The condition 2.7 is

important to guarantee the material stability of the calculated critical point, because

the solutions are yielded by equations 2.5 and 2.6 may represent unstable situations

which cannot be observed experimentally. Omitting condition 2.7 is not usually a

serious oversight for vapour -liquid critical properties because physically unrealistic
22
results can usually be detected by inspection, for example, by observing an

anomalous trend in the predicted critical volume. Similarly, there are ample

experimental precedents for binary mixtures to confidently identify physically

erroneous solutions of equations 2.5 and 2.6. In addition, a critical point must satisfy

the conditions for mechanical stability (Sadus, 1992a):

p
<0
V T

2
2A 2A 2A
2 2 > 0 (2.8)

1 T ,V
x V
T 1 T ,V
x V

2A
2 > 0
x1 T ,V

2.4 Nonclassical Critical Phenomena

Critical behaviour is called classical if the Helmholtz function can be expanded at

the critical point in terms of dV = V Vc and dT = T Tc. Where V is the volume, and

T is the temperature of the fluid. The subscript c denotes the critical values. However,

it is well known (Levelt Sengers et al., 1986) that the critical state cannot be treated

analytically. It is increasingly apparent that the nonclassical nature of the critical state

must be taken into account to genuinely improve the prediction of critical or near

critical equilibria. It can be argued that the use of optimising parameters is at least,

partly due to an inadequate description of the critical state. On the other hand, the

predictive capabilities of current nonclassical models of the critical state are very

restricted.

23
The variation of the physical properties of fluids such as specific heat and isothermal

compressibility approaching the critical point has been extensively investigated by

Stanley (1971), Levelt Sengers (1983) and Levelt Sengers et al. (1983). These studies

can be used to define critical exponents, which provide an insight into the nature of

the critical transition. Consider a function f(e), where e 1 T / Tc is a dimensionless

variable that reflects the difference in temperature from the critical temperature. It is

assumed that f(e) is both positive and continuous for sufficiently small positive values

of e and that the following limit exists:

lim [ ln f ( ) / ln ] (2.9)
0

The limit ? is the critical point exponent associated with the f (e) function. Four of the

most common critical exponents of a pure substance are defined in Table 2.1. In Ta ble

2.1, we also give values of these critical exponents which are typically obtained using

the van der Waals equation of state or other mean field theories, and compare them to

experimental values. The discrepancy between theory and experiment indicates that

the critical state cannot be treated analytically. Analogous critical exponents of binary

mixtures can also be defined (Scott, 1972; Rowlinson and Swinton, 1982; Kiselev and

Rainwater, 1997; Jiang and Prausnitz, 1999; Agayan et al. 2001).

Extrapolations of careful experimental measurements in the region of the critical point

of binary fluid mixtures (Scott, 1972) indicate that similar non-classical exponents are

required for binary critical phenomena as for pure component near critical behaviour.

The most common critical exponents of binary mixtures are also defined in Table 2.1.

For classical behaviour of fluid mixtures, the Helmholtz function with x as the mole

24
fraction also can be expanded at the critical line in terms of its independent variables.

The implications of classical behaviour can be visualized by assuming with van der

Waals, that the mixture at constant composition x is in corresponding states with it

components.

Table 2.1 Definition and values of Critical exponents for one-component and binary mixtures
fluid.

Definition
Property Classical Experimental
Pure Component Binary mixtures Value Value

Isochoric heat capacity Cv ~ | T - Tc |-a Cpx ~ | T - Tc | -a a=0 a = 0.080.15

Saturation ( ? ? ) ~ ( Tc T ) (x x) ~ ( Tc T ) = 0.5 = 0.32 0.36

Isothermal compressibility KT ~ | T - Tc |-? KT ~ | T - Tc |-? ?=1 ? = 0.96 1.28

Pressure | p p c | ~ | ? ?c | d | p p c | ~ | xxc | d d=3 d = 4.35.1

The scaling laws (Sengers and Levelt Sengers, 1978) that describe the

thermodynamics of near-critical fluids can be viewed as a compact packaging of the

table 2.1. They imply the exponent equalities

2 = ( + 1); = ( 1) (2.10)

So that only two exponents are independent. Scaling laws will be given in Chapter 3.

Griffiths and Wheeler (1970) developed an alternative to the conventional description

of fluid properties based on a division of variables into fields and densities. This

formalism has been applied to the detection of tricritical points in quasi-binary

mixtures (Pegg et al., 1983). A scaling approach is frequently employed to

25
incorporate the effects of nonclassical behaviour. Leung and Griffiths (1973)

constructed a scaled thermodynamic potential for vapour-liquid equilibria in binary

mixtures by spitting the potential into a regular and scaled part. Rainwater and

Moldover (1983) proposed a modified version of the Leung-Griffiths equation. For

appropriately chosen isomorphic variables, the thermodynamic potential of binary

mixtures has the same form as the thermodynamic potential of a one component fluid

(Fisher, 1982), and scaling law calculations for vapour-liquid equilibrium of a wide

variety of binary mixtures have been developed (Rainwater, 1992). The asymptotic

singular critical behaviour of the thermodynamic properties can be described in terms

of scaling laws with common critical exponents and common scaling function

(Anisimov and Kiselev, 1992).

2.5 Fluid Models

To calculate critical phenomena we need an expression for the Helmhotz function. In

this work we have used conformal solution theory but often approaches based on

perturbation theory or using the equation of state directly are also possible. We will

introduce conformal solution and perturbation theories in this section. Equations of

state will be discussed in Chapter 3.

2.5.1 Conformal Solution Theory

Conformal solution theory was first proposed by Longuet-Higgins (1950), for

mixtures in which the intermolecular forces between molecules are similar. It was

developed by Brown (1957), Hicks and Young, (1975) and Shukla et al. (1986), and it

26
has been used for the prediction of critical phenomena of binary mixtures. It predicts

mixture properties from the experimentally measured properties of one or more pure

reference substances. This theory assumes that the molecules are conformal, and was

originally devised for nonpolar, spherical molecules of similar size, but it has been

successfully applied to a much more diverse range of mixtures (Toczylkin and Young,

1980; Sadus and Young, 1985; Christou et al., 1986). The van der Waals one -fluid

(Scott, 1956), two-fluid (Leland et al., 1969), and three-fluid (Scott, 1956) models are

examples of fluid models used with conformal solution theory. The predictions of

conformal solution theory compare very favourably with the results of computer

simulation of phase equilibria of mixtures of molecules that differ vastly in volume

(Harismiadis et al., 1991). The works of Sadus and Young (1987, 1988) and Sadus

(1992a, b) demonstrate that conformal solution theory is also useful for the prediction

of critical phenomena of ternary mixtures. Wei (1998) extensively reviewed

conformal solution and perturbation theories.

2.5.1.1 The Helmholtz Function

The principle of corresponding states (Brown, 1957, Rowlinson, 1969, Shukla et al.

1986) is fundamental to conformal solution theory. It enables us to relate the

intermolecular potential of one pure substance to that of another. The Helmholtz

function of a pure substance (1) to the Helmholtz function of a reference substance (0)

can be related by

A1* (V , T ) = f 11 A0* (V / h11, T / f 11 ) RT ln h11 (2.11)

27
where A1 * and A0 * denote the configurational Helmholtz function of the pure fluid and

the reference fluid respectively, and f 11 and h11 are the conformal parameters,

h11 = g113
(2.12)
f 11 = 11 / 00

where g 11 = 11 / 00 , and terms are the characteristic energy and distance of an

interaction between two molecules.

2.5.1.2 One-Fluid Model

The simplest form of conformal solution theory is the van der Waals one-fluid model

(Scott, 1956; Leland et al., 1968; Henderson and Leonard, 1971). In the model, a real

mixture is considered to be a single hypothetical pure fluid. Mixture properties are

expressed in the terms of the equation of state of a pure fluid with composition

dependent size and energy parameters. In the van der Waals one-fluid model, the

Helmholtz function Am of the mixture can be approximated by that of a hypothetical

pure substance, the equivalent substance (subscript es).

The Helmholtz function for the binary mixtures Am is obtained (Hicks and Young,

1975, Sadus, 1992a) from:

Am = Aes* + Acb (2.13)

28
The residual Helmholtz function Aes* is represented:

Aes* = f es A0* (V / hes , T / f es ) RT ln hes (2.14)

and the combinatorial Helmholtz energy of the mixture Acb has the value:

Acb = RT x i ln x i (2.15)
i

where R is the universal gas constant, A0* is the configurational Helmholtz function of

the reference substance. The configurational contribution can be obtained from an

equation of state. The f and h terms are conformal parameters which can be deduced

from the critical temperatures and volume, respectively. Normally, one of the

components of the mixture is chosen as the reference (denoted by the subscript 0) and

the conformal properties of the other pure substances are obtained relative to it, for

example, f 11 = T11c / T00c ; h11 = V11c / V00c etc.

The conformal parameters for the equivalent substance of binary mixtures are

evaluated from the van der Waals prescriptions

f es hes = x x
i =1, 2 j =1 , 2
i j f ij hij (2.16)

hes = x x h
i =1 , 2 j =1, 2
i j ij (2.17)

29
The contribution of unlike interactions f ij, hij ( i j ) are typically obtained from

combining rules. Combining rules are discussed in Chapter 3.

2.1.5.3 Two or More Fluid Model

We have used the one-fluid model in our work, but we will briefly discuss multi-fluid

models here.

The averaging for the two-fluid model (Leland et al., 1969) is made partly on the

intermolecular interactions, and partly on the configurational Helmholtz energy. The

average intermolecular interaction for both molecular species is calculated and used to

generate the Helmholtz function Am of the mixture.

Am = (1 x )[ f 1 A0 (V / h1 , T / f 1 ) RT ln h1 ] + x[ f 2 A0 (V / h2 , T / f 2 ) RT ln h2 ] (2.18)

Where the average reducing ratios for species i, fi and hi are usually functions of

composition.

In the van der Waals three-fluid model (Scott, 1956), there is no preliminary

averaging of the conformal parameters to produce equivalent substances and the

Helmholtz Am of the mixtures is obtained directly from the set of f ij and hij ,

Am = (1 x ) 2 A11 + 2 x (1 x ) A12 + x 2 A22 (2.19)

and

30
Aij (V , T ) = f ij A0 (V / hij , T / f ij ) RT ln hij (2.20)

The n-fluid models were tested against computer simulation result by McDonald

(1972, 1973), Henderson (1974a), and Shing and Gubbins (1982, 1983). The results

showed that the van der Waals one -fluid model was superior to the two-fluid and

three-fluid model (Wei, 1998). Hicks (1976) used the two-fluid model to calculate

high pressure phase equilibria and critical properties of binary mixtures and found that

the two-fluid model suffers from internal inconsistencies in the critical region and

concluded that the three-fluid model suffers from the same defect as the two-fluid

model. Rowlinson and Swinton (1982) also concluded that the two-fluid model is

more difficult to use in practice and the three-fluid model imputes an unrealistically

high degree of ordering to a dense fluid system. Therefore, the two-fluid and three-

fluid models have not been widely used. Recently, Moore et al. (1995) applied the van

der Waals one-fluid to model the temperature dependency of infinite dilution activity

coefficients. Wei and Sadus (1999) used conformal solution theory in conjunction

with the one-fluid model and the Guggenheim equation of state to calculate the

critical properties of binary and ternary mixtures characterised by molecules of

identical size but different energy parameters. Brandt et al. (2000) also applied the

conformal solution theory in conjunction with the one -fluid model and the

Guggenheim equation of state to supplement the experimental measurements.

2.5.1.4 Hard-Sphere Expansion

Another conformal solution model is represented by the Hard-Sphere Expansion

(HSE) conformal solution theory (Mansoori and Leland, 1972). This model uses the

explicitly known properties of a hard sphere mixture for modelling the repulsive

31
interactions, and the attractive contributions were taken into account by a one -fluid

model.

f 2 h 2 3 = xi x j ij2 ij3 (2.21)

i j

fh 3 = x i x j ij ij3 (2.22)
i j

Where xi and xj are the mole fraction of the component i and j, the s is the molecular

diameter, and the double summation is over all the components of the mixture.

Mansoori and Leland (1972) tested the HSE model by doing a series of comparisons

between the predicted excess properties on mixing and these properties from

experimental and Monte Carlo data. Chang et al. (1979) developed a general method

of prediction the effective molecular diameters and the thermodynamic properties for

fluid mixtures based on the HSE model. Shukla et al. (1986) compared various forms

of conformal solution theories, including HSE model, with computer simulation

results for Lennard-Jones mixtures under different conditions. Dahmani and Wichterle

(1999) presented a modification of the HSE mode with the polarity of molecules for

prediction of vapour-liquid equilibrium at constant temperature in binary, ternary and

quaternary mixtures consisting of chlorobutanes and C-7 hydrocarbons.

32
2.5.2 Perturbation Theories

Perturbation theories are a commonly used alternative to conformal solution theory

(Wei, 1998). In perturbation theory, the Helmholtz function is determined from a

reference system plus perturbation terms, which are obtained from the radial

distribution function of the reference system. Different reference terms can be used to

account for different fluid properties such as molecular size, shape or polarity. The

perturbation theory approach is based on the assumption that the structure of a fluid is

primarily determined by strong, short range repulsive forces and that the structure of a

fluid is only slightly modified by weak, long range attractive forces. Therefore, in the

perturbation theories for spherical molecules, the hard sphere fluid is taken as the

reference and the weak attractive forces are taken as the perturbation. For example,

Barker and Henderson (1967) considered the long-ranged attractive contributions can

be perturbations, they determined the Helmholtz function form:

2
1 1
A = A0 + A1 + A2 + .... (2.23)
kT kT

where k is Boltzmanns constant, T is temperature, A0 is the Helmholtz function of the

reference system, A1 and A2 are the first and second order perturbation therms,

respectively.

This approach has been exploited in different forms of perturbation theories of pure

fluids (Barker and Henderson, 1967; Weeks et al., 1971; Verlet and Weis, 1972) and

fluid mixtures (Mansoori and Leland, 1970; Grundke et al., 1973; Shukla, 1987; Lotfi

and Fischer, 1989; Fotouh and Shukla, 1996a). Furthermore, Boublik (1976, 1987,

33
1988) proposed a convex molecule perturbation theory based on the Barker-

Henderson (1967) perturbation theory. For associating fluids, Wertheim (1984a, b,

1986a, b) developed his thermodynamic perturbation theory.

Perturbation theories offer a possible improvement (Gubbins, 1983) over van der

Waals n-fluid theories when the molecules differ in size. Fotouh and Shukla (1996a,

b, 1997) compared van der Waals one-fluid theory and perturbation theory, and

obtained results, which showed that perturbation theory is better than van der Waals

one-fluid model for predicting thermodynamic properties when the size of molecules

are different. Detailed discussion of perturbation theories were given in several

reviews (McDonald, 1973; Henderson, 1974a; Barker and Henderson, 1976; Gray and

Gubbins, 1984; Fotouh and Shukla, 1997).

Rogers and Prausnitz (1971) applied Eq. (2.23) with good success in the study of high

pressure vapour-liquid equilibria in mixtures. Henderson (1974b) applied the Barker-

Henderson theory to mixtures of squa re-well molecules simulated and obtained good

results. By extending the perturbation expansion of Barker and Henderson (1967),

Vimalchand and Donohue (1985) and Vimalchand et al. (1986) developed the

Perturbed Anisotropic Chain Theory (PACT). Boublik (1976) proposed a simple

perturbation method for systems of interacting convex molecules to have an extension

of the Barker and Henderson theory via the Kihara pair potential (Kihara, 1963).

Boublik (1990) applied the convex molecule perturbation theory to describe the two-

phase behaviour of 21 pure nonpolar compounds with fair accuracy. Pavlicek and

Boublik (1992) reported parameters of the Kihara generalised pair potential for n-

alkanes, branched alkanes, and 1-chloroalkanes, all modelled as rod-like molecules.

Pavlicek et al. (1995) found that the convex molecule perturbation theory represents a

34
suitable tool for calculating the equilibrium thermodynamic properties of real pure

nonpolar nonspherical fluids along their vapour -liquid coexistence region. Boublik

(1997) also extended the perturbation theory to ternary mixtures of polar non-

spherical molecules.

The Weeks-Chandler-Andersen model (Weeks et al., 1971) is another widely cited

perturbation theory. The model also divides the intermolecular (pair) potential into a

reference system pair potential part and a perturbation potential part. However, in the

Weeks-Chandler-Andersen model the reference system pair potential part includes all

the repulsive forces in the Lennard-Jones potential, and the perturbation potential part

includes all the attractions. The Weeks-Chandler-Andersen theory has been extended

to fluids with linear molecules (Kohler et al., 1979), triangular, tetrahedra and

octahedral molecules (Lustig, 1986, 1987) and to mixtures (Fischer and Lago, 1983).

Shukla et al. (1986) combined the Lee and Levesque (1973) theory with the Grundke

and Henderson (1973) theory of pair correlation function and presented an improved

form of the Lee and Levesque theory, called WCA-LL-GH theory. Shukla (1987)

presented an improved form of the first-order perturbation theory using some

modifications in WCA-LL-GH. Based on these modifications, Fotouh and Shukla

(1996b) presented another improved first-order hard sphere perturbation theory,

which consists of the first-order perturbation theory of high temperature

approximation (Shukla, 1987) and the random phase approximation (Andersen et al.,

1972).

The work of Scalise et al. (1987) is an example of the application of a perturbation

theory to critical phenomena. Typically, perturbation models determine the Helmholtz

function from:

35
A = Aiso + A2 + A3 + ... (2.24)

where Aiso is the angle-independent isotropic contribution, and the A2 , A3 terms are

the pair and triplet correlation functions, respectively. The above series was obtained

by applying the Pad approximation:

A = Aiso + A2 /(1 A3 / A2 ) (2.25)

Scalise et al. (1987) specifically calculated critical phenomena of fluids with

permanent quadrupole moments.

Wertheim (1984a, b) developed a thermodynamic perturbation theory (TPT) theory to

explain the behaviour of fluids, which have short-range directional attractive

interactions like those in associating or hydrogen-bonding fluids. Weitheim (1987)

also extended his theory to chain fluids and developed the first-order and second-

order thermodynamic perturbation theories (TPT1 and TPT2) for a polydisperse

mixture of chains of varying lengths. Chapman et al. (1988) and Jackson et al. (1988)

also extended Wertheims theory to mixtures of spheres and of chain molecules and

tested the theory against Monte Carlo simulation. This will be discussed in greater

detail in Chapter 3, in the context of equations of state for hard chains.

2.6 Calculation Procedures for Binary Mixtures

To calculate phase equilibria, an effective numerical technique is required. There are

many techniques (Henrici, 1964; Traub, 1964; Acton, 1970; Hicks and Young, 1977;

36
Press et al, 1988) available for equations of two or three variables. We will introduce

two calculation methods.

2.6.1 Newton-Raphson Method

A common method involving successive refinement of an initial approximate solution

until convergence is obtained. For isolated n-dimensional root geometries combined

with starting values that are noticeably nearer to one root than any of the others,

almost any reasonable method will converge to the root. If explicit first derivatives are

calculable, the Newton-Raphson method (Henrici, 1964; Acton, 1970; Walas, 1985;

Press et al, 1988) is the method of first choice. It is straightforward and converges

quadratically. The application of the Newton-Raphson method to the critical

conditions, allows us to determinate the critical temperature and volume from:

V i+1 = V i {X i (W / T ) i W i (X / T ) i } / J (2.26)

T i +1 = T i {W i (X / V ) i X i ( W / V ) i } / J (2.27)

J = ( W / T ) i (X / V ) i ( W / V ) i ( X / T ) i (2.28)

Although the Newton-Raphason method has been widely used for critical

calculations, it is largely restricted to the calculation of vapour-liquid critical

properties, but reliable estimates for liquid-liquid critical points cannot generally be

obtained a priori. Consequently, the use of the Newton-Raphson method is restricted

37
to type I and II phenomena. To predict critical points of other types of binary

mixtures, a more general solution is required (Sadus, 1994).

2.6.2 Hicks-Young Algorithm

In this work, the Hicks and Young algorithm (Hicks and Young, 1977, Sadus and

Young, 1987) is used to locate all types of critical points. The algorithm works by

tracking either Eq. (2.3) and (2.4) at a fixed composition and simultaneously by

monitoring the sign of the other function in a temperature and volume search area. It

is computationally easier to trace W = 0 rather than X = 0, because the second

derivatives of chemical potential is of increasing complexity in the latter condition.

The search procedure is illustrated in Figure 2.4. Two points 1 and 6 are designated at

a small interval on either side of the entry point, which lies on one side of a square

defined by the points 2 to 5. The sign of W is compared between the adjacent pairs of

points: 1-2; 2-3; 3-4; 4-5; and 5-6. A change of sign indicates that the condition W = 0

was passed between the pair of points, and the times is odd number. The sign of X is

checked at these points. If the sign of X also has changed, then an intersection of W =

0 and X = 0 has been passed. The search procedure is scaled down until the critical

point is accurately found. If the sign of X is unchanged, then the new point is used to

establish the direction of the W = 0 curve and the next search square is arranged.

The major advantages of the Hicks-Young algorithm are:

1. In principle, all solutions of the critical conditions are located.

2. It does not require initial approximations.

3. No additional derivatives are required.

38
The first and second aspects are very important for the a priori prediction of critical

equilibria. Van Pelt (1992) and Castier and Sandler (1997a, b) employed the Hicks

and Young algorithm to perform critical point calculations in binary systems.

Recently, Wang et al. (2000) applied the algorithm to calculate closed-loop liquid-

liquid equilibria. A disadvantage of the algorithm is that it is much more

computational expensive than simple iterative approaches. However, the ability to

confidently locate all possible solutions to the critical conditions out weighs this

disadvantage. This is particularly important for the calculation of global phase

diagrams (Chapter 5 & 6).

W=0

X=0

2 3

5 4

V
Figure 2.4. Hicks -Young algorithm for locating the critical point.

39
References

Acton, F. S. (1970). Numerical Methods That Work, Harper and Row, New York.

Agayan, V. A., Anisimov, M. A. and Sengers, J. V. (2001). Crossover Parameter

Equation of State for Ising-Like Systems, Phys. Rev. E 64, 6125-6144.

Andersen, H. C., Chandler, D. and Weeks, J. D. (1972). Roles of Repulsive and

Attractive Forces in Liquids: the Optimized Random Phase Approximation, J.

Chem. Phys., 56, 3812-3823.

Anisimov, M. A. and Kiselev, S. B. (1992). Universal Crossover Approach to

Description of Thermodynamic Properties Fluids and Fluid Mixtures, Taylor &

Francis, New York.

Barker, J. A. and Henderson, D. (1967). Perturbation Theory and Equation of State

for Fluids. II. A Successful Theory of Liquids, J. Chem. Phys. 47, 4717-4721.

Barker, J. A. and Henderson, D. (1976). What is Liquid? Understanding the States of

Matter, Rev. Mod. Phys., 48, 587-671.

Beegle, B. L., Modell, M. and Reid, R. C., (1974). Thermodynamic Stability Criterion

for Pure Substances and Mixtures, AIChE J., 20, 1194-1206.

Boublik, T. (1976). Perturbation Theory for Fluids of Rod-Like Molecules Interaction

via the Kihara Potential, Mol. Phys., 32, 1737-1749.

Boublik, T. (1987). Simple Perturbation Method for Convex-Molecule Fluids, J.

Chem. Phys., 87, 1751-1756.

Boublik, T. (1988). Simple Perturbation Method for Convex Molecule Mixtures, J.

Phys. Chem. , 92, 2629-2633.

Boublik, T. (1990). Equilibrium Behaviour of Pure Substances from the Convex

Molecule Perturbation Theory, Fluid Phase Equilib., 54, 221-235.

Boublik, T. (1992). Perturbation Theory for Mixtures of Polar and Non-Polar

Anisotropic Molecules, Fluid Phase Equilib., 73, 211-224.

40
Boublik, T. (1997). Phase Equilibria of Systems Considered in Supercritical Fluid

Extraction Determined from Perturbation Theory, Mol. Phys., 90, 585-591.

Brandt, E., Franck, E. U., Wei, Y.S., Sadus, R. J., (2000). Phase Behaviour of Carbon

Dioxide-Benzene-Water Ternary Mixtures at High Pressures and Temperatures up

to 300 MPa and 600K, Phys. Chem. Chem. Phys., 18, 4157-4164.

Brown, W. B. (1957). The Statistical Thermodynamics of Mixtures of Lennard-Jones

Molecules I. Random Mixtures, Philos. Trans. Roy. Soc. London A, 250, 175-220.

Bumstead, H. A. and van Names, R. G., (1906). Eds. The Scientific Papers of J.

Willard Gibbs: I. Thermodynamics, Longmans, New York.

Castier, M. and Sandler, S. I. (1997a). Critical Points with the Wong-Sandler Mixing

Rule. I. Calculations with the van der Waals Equation of State, Chem. Eng. Sci.,

52, 3393-3399.

Castier, M. and Sandler, S. I. (1997b). Critical Points with the Wong-Sandler Mixing

Rule. II. Calculations with a Modified Peng-Robinson Equation of State, Chem.

Eng. Sci., 52, 3579-3588.

Chang, J. I. C., Hwu, F. S. S. and Leland, T. W. (1979). Effective Molecular

Diameters for Fluid Mixtures, In Chao K. C. and Robinson, Jr. R. L. (Eds),

Equations of State in Engineering and Research, Advances in Chemistry Series,

182, American Chemical Society, Washington.

Chapman, W. G., Jackson, G. and Gubbins, K. E. (1988). Phase Equilibria of

Associating Fluids: Chain Molecules with Multiple Bonding Sites, Mol. Phys., 65,

1057-1079.

Christou, G., Morrow, T., Sadus, R. J. and Young C. L. (1986). Phase Behaviour of

Fluorocarbon + Hydrocarbon Mixtures: Interpretation of Type II and Type III

Behaviour in Terms of a Hard sphere + Attractive Term Equation of State,

Fluid Phase Equilib., 25, 263-272.

41
Dahmani O. and Wichterle I. (1999). Hard Sphere Expansion Conformal Solution

Theory with the Lennard-Jones (12-6) Intermolecular Potential, Fluid Phase

Equilib., 161, 21-32.

Donnan, F. G. and Hass, A., (1936). Eds., Commentary on the Scientific Writings of

J. Willard Gibbs: 1. Thermodynamics, Yale Univ. Press, New Haven.

Fisher, M. E. (1982). Hahne, F. J. W. (Ed.), Critical Phenomena. Lecture Notes in

Physics, Vol. 186, Spr inger, Berlin.

Fischer, J. and Lago S. (1983). Thermodynamic Perturbation Theory for Molecular

Liquid Mixtures, J. Chem. Phys., 78, 5750-5758.

Fotouh, K. and Shukla, K., (1996a). Thermodynamic Properties of Ternary Fluid

Mixtures from the Improved Perturbation Theory and van der Waals One-Fluid

Theory. I. Model Mixtures, Chem. Eng. Sci., 51, 4923-4931.

Fotouh, K. and Shukla, K., (1996b). Thermodynamic Properties of Ternary Fluid

Mixtures from the Improved Perturbation Theory and van der Waals One-Fluid

Theory. II. Real Mixtures, Chem. Eng. Sci., 51, 4933-4937.

Fotouh, K. and Shukla, K., (1997). An Improved Perturbation Theory and van der

Waals One-Fluid Theory of Binary Fluid Mixtures. 1. Total and Excess

Thermodynamic Properties, Fluid Phase Equilib., 127, 45-70.

Gibbs, W. J., (1928). The Collected Works of J. Willard Gibbs: 1. Thermodynamics,

Longmans, New York.

Gray, C. G. and Gubbins, K. E. (1984). Theory of Molecular Fluids, Oxford

University Press.

Grundke, E. W., Henderson, D., Barker, J. A. and Leonard, P. J. (1973). Perturbation

Theory and the Excess Properties of Mixtures of Simple Liquids, J. Phys., 25,

883-896.

42
Gubbins, K. E., Shing, K. S. and Streett, W. B. (1983). Fluid Phase Equilibria:

Experiment, Computer Simulation, and Theory, J. Phys. Chem., 87, 4573-4585.

Harismiadis, V. I., Koutras, N. K., Tassios, D. P. and Panagiotopoulos, A. Z. (1991).

How Good is Conformal Solutions Theory for Phase Equilibrium Predictions?;

Gibbs Ensemble Simulations of Binary Lennard-Jones Mixtures, Fluid Phase

Equilib., 65, 1-18.

Heidemann, R. A., (1975). The Criteria for Thermodynamic Stability, AIChE J., 21,

824-826.

Henderson, D. and Leonard P. J. (1971). In Eyring, H., Henderson, D. and Jost, W.

(Eds.), Physical Chemistry, An Advanced Treatise, Vol. 8B: Liquid State,

Academic Press, London.

Henderson, D. (1974a). Theory of Simple Mixtures, Ann. Rev. Phys. Chem., 25, 461-

483.

Henderson, D. (1974b). Perturbation Theory for a Mixture of Hard Spheres and

Square-Well Molecules, J. Chem. Phys., 61, 926-931.

Henrici, P. (1964). Elements of Numerical Analysis, Wiley, New York.

Hicks, C. P. and Young C. L. (1975). The Gas-liquid Critical Properties of Binary

Mixtures, Chem. Rev., 75, 119-175.

Hicks, C. P. (1976). N-Fluid Models and Phase Equilibrium, J. Chem. Soc. Faraday

Trans. II., 72, 423-425.

Hicks, C. P. and Young C. L. (1977). Theoretical Predictions of Phase Behaviour at

High Temperatures and Pressures for Non-Polar Mixtures: 1. Computer Solution

Techniques and Stability Tests, J. Chem. Soc., Faraday II, 73, 597-612.

Jackson, G., Chapman, W. G. and Gubbins, K. E. (1988). Phase Equilibria of

Associating Fluid: Spherial Molecules with Multiple Bonding Sites, Mol. Phys.,

65, 1-31.

43
Jiang, J. and Prausnitz, J. M., (1999). Equation of State for Thermodynamic

Properties of Chain Fluids Near-To and Far-From the Vapour-Liquid Critical

Region, J. Chem. Phys., 111, 5964-5974.

Kihara, T. (1963). Convex Molecules in Gaseous and Crystalline States, Adv. Chem.

Phys., 5, 147-188.

Kiselev, S. B. and Rainwater, J. C., (1997). Extended Law of Corresponding States

and Thermodynamic Properties of Binary Mixtures in and Beyond the Critical

Region, Fluid Phase Equilib., 141, 129-154.

Kohler, F., Quirke, N. and Perram, J. W. (1979). Perturbation Theory with a Hard

Dumb-bell Reference System, J. Chem. Phys., 71, 4128-4131.

Lee, L. L. and Levesque, D. (1973). Perturbation Theory for Mixtures of Simple

Liquids, Mol. Phys., 26, 1351-1370.

Leland, T. W., Rowlinson, J. S. and Sather, G. A. (1968). Statistical Thermodynamics

of Mixtures of Molecules of Different Sizes, Trans. Faraday Soc., 64, 1447-1460.

Leland, T. W., Rowlinson, J. S., Sather, G. A. and Watson, I. D. (1969). Statistical

Thermodynamics of Two-Fluid Models of Mixtures, Trans. Faraday Soc., 65,

2034-2043.

Levelt Sengers, J. M. H. (1983). The State of the Critical State of Fluids, Pure &

Appl. Chem., 55, 437-441.

Levelt Sengers, J. M. H, Morrison, G. and Chang, R. F. (1983). Critical Behaviour in

Fluids and Fluid Mixtures, Fluid Phase Equilib., 14, 19-44.

Levelt Sengers, J. M. H, Chang, R. F. and Morrison, G. (1986). Nonclassical

Description of (Dilute) Near-Critical Mixtures, ACS Symp. Ser., 300, 110-123.

Longuet-Higgins, H. C., (1950). A Theory of the Origin of Microseisms, Philos.

Trans. Roy. Soc. London A, 243, 1-35.

44
Lotfi, A. and Fischer, J. (1989). Chemical Potientials of Model and Real Dense Fluid

Mixtures from Perturbation Theory and Simulations, Mol. Phys., 66, 199-219.

Lustig, R. (1986). A Thermodynamic Perturbation Theory for Non-linear Multicentre

Lennard-Jones Molecules with an Anisotropic Reference System, Mol. Phys., 59,

173-194.

Lusting, R. (1987). Application of Thermodynamic Perturbation Theory to

Multicentre Lennard-Johnes Molecules. Results for CF4 , CCl4 , neo-C5 H12 and SF6

as Tetrahedral and Octahedral Models, Fluid Phase Equilib., 32, 117-137.

Mansoori, G. A. and Canfield, F. B. (1970). Inequalities for the Helmholtz Free

Energy, J. Chem. Phys., 53, 1618-1619.

Mansoori, G. A. and Leland, Jr. T. W. (1972). Statistical Thermodynamics of

Mixtures. A New Version for Theory of Conformal Solution, J. Chem. Soc.

Faraday Trans. II, 68, 320-344.

McDonald, I. R. (1972). NPT-Ensemble Monte Carlo Calculations for Binary Liquid

Mixtures, Mol. Phys., 23, 41-58.

McDonald, I. R. (1973). In: K. Singer (Ed), Statistical Mechanics, Vol. 1. Chemical

Society, London.

Modell, M., (1977). Criteria of Criticality, ACS Symp. Ser., 60, 369-373.

Moore, R.C., Jonah, D., Cochran, H. D., and Bienkowski, P.R., (1995). Modeling

Infinite Dilution Activity Coefficients of Environmental Pollutants in Water Using

Conformal Solution Theory, Separation. Sci. & Tech., 30,1981-1996.

Pavlicek, J., Aim, K. and Boublik, T. (1995). Fluid of the Kihara Molecules. 1. Pair

Potential Parameters of n-Alkanes from the Vapour-Liquid Coexistence Data, J.

Phys. Chem., 99, 15662-15668.

45
Peng, D. -Y. and Robinson, D. B. (1977). A Rigorous Method for Predicting the

Critical Properties of Multicomponent Systems from an Equation of State, AIChE

J., 23, 137-144.

Press, W. H. (1988). Numerical Recipes in C. The Art of Scientific Computing,

Cambridge University Press, Cambridge.

Rainwater, J. C. (1992). Supercritical Fluid Technology, In: J. F. Ely, T. J. Bruno

(Eds.), CRC Press, Florida.

Reid, R. C. and Beegle, B. L., (1977). Critical Point Criteria in Legendre Transform

Notation, AIChE J., 23, 726-732.

Rogers, B. L. and Prausnitz, J. M. (1971). Calculation of High-Pressure Vapour-

Liquid Equilibriums with a Perturbed Hard-Sphere Equation of State, Trans.

Faraday Soc., 67, 3474-3483.

Rowlinson, J. S. (1969). Liquid and Liquid Mixtures, 2nd Ed., Butterworths, London.

Rowlinson, J. S. and Swinton, F. L. (1982). Liquids and liquid mixtures, 3rd Ed.,

Butterworths, London.

Sadus, R. J. and Young, C. L. (1985). Phase Behaviour of Binary n-Alkanenitrile and

n-Alkanes and Nitromethane and Alkane Mixtures, Fluid Phase Equilib., 22, 225-

229.

Sadus, R. J. and Young, C. L. (1987). The Critical Properties of Ternary Mixtures:

Hydrocarbon, Acetone and Alkanenitrile Mixtures, Chem. Eng. Sci., 42, 1717-

1719.

Sadus, R. J., Young, C. L. and Svejda, P. (1988). Application of Hard Convex Body

and Hard Sphere Equation of State to the Critical Properties of Binary Mixtures,

Fluid Phase Equilib., 39, 89-99.

Sadus, R. J. (1992a). High Pressure Phase Behaviour of Multicomponent Fluid

Mixtures, Elsevier, Amsterdam.

46
Sadus, R. J. (1992b). Novel Critical Transitions in Ternary Fluids Mixtures, J. Phys.

Chem., 96, 5197-5202.

Sadus, R. J. (1994) Calculating Critical Transitions of Fluid Mixtures: Theory Vs.

Experiment, AIChE J. 40, 1376-1403.

Scalise, O. H., Zarragoicoechea, G. J., Rodriguez, A. E. and Gianotti, R. D. (1987).

Quadrupolar Binary Fluid Mixtures Critical Lines From a Hard-Sphere Lennard-

Jones Quadrupolar Molecular Model, J. Chem. Phys., 86, 6432-6437.

Scott, R. L. (1956). Corresponding State Treatment of Nonelectrolyte Solutions, J.

Chem. Phys., 25, 193-205.

Scott, R. L. (1972). The Thermodynamics of Critical Phenomena in Fluid Mixtures,

Ber. Bunsenges. Phys. Chem., 76, 296-315.

Sengers, J. V. and Levelt Sengers, J. M. H. (1978). In: Progress in Liquid Physics,

Croxton, C. A. (Ed.), Wiley: Chichester.

Shing, K. S. and Gubbins, K. E. (1982). The Chemical Potiential in Dense Fluids and

Fluids Mixtures via Computer Simulation, Mol. Phys., 46, 1109-1128.

Shing, K. S. and Gubbins, K. E. (1983). The Chemical Potiential in Non-Ideal liquid

Mixtures. Computer Simulation and Theory, Mol. Phys., 49, 1121-1138.

Shukla, K. P., Luckas, M., Marquardt, H. and Lucas, K. (1986). Conformal Solutions:

Which Model for Which Application?, Fluid Phase Equilib., 26, 129-147.

Shukla, K. P. (1987). Thermodynamic Properties of Simple Fluid Mixtures from

Perturbation Theory, Mol. Phys., 62, 1143-1163.

Spear, R. R., Robinson, R. L., Jr., and Chao, K. -C. (1971). Critical States of Ternary

Mixtures and Equations of State, Ind. Eng. Chem. Fundam., 10, 588-592.

Spear, R. R., Robinson, R. L., Jr., and Chao, K. -C. (1969). Critical States of Mixtures

and Equations of State, Ind. Eng. Chem. Fundam., 8, 2-8.

47
Stanley, H. E. (1971). Introduction to Phase Transitions and Critical Phenomena,

Clarendon Press, Oxford.

Traub, J. J. (1964). Iterative Methods of the Solution of Equations, Prentice Hall,

Englewood Cliffs, New Jevsey.

Toczylkin, L. S. and Young, C. L. (1980). Gas-Liquid Critical Temperatures of

Mixtures Containing Electron Donors. Part 1. Ether Mixtures, J. Chem.

Thermodyn., 12, 355-364.

Van Konynenburg, P. H. and Scott, R. L., (1980). Critical Lines and Phase Equilibria

in Binary van der Waals Mixtures, Philos. Trans. Roy. Soc. London A, 298, 495-

540.

Verlet, L. and Weis, J. J. (1972). Equilibrium Theory of Simple Liquids, Phys. Rev. A,

5, 939-952.

Walas, S. M., (1985). Phase Equilibria in Chemical Engineering, Butterworths,

Boston.

Wang, J. L., Wu, G. W. and Sadus, R. J. (2000). Closed-Loop Liquid-Liquid

Equilibria and the Global Phase Behaviour of Binary Mixtures Involving Hard-

Sphere + van der Waals Interactions, Mol. Phys. 98, 715-723.

Weeks, J. D., Chandler, D. and Andersen, H. C. (1971). Role of Repulsive Forces in

Determining the Equilibrium Structure of Simple Liquids, J. Chem. Phys., 54,

5237-5247.

Wei, Y. S. and Sadus, R. J., (1999). Phase Behaviour of Ternary Mixtures: a

Theoretical Investigation of the Critical Properties of Mixtures with Equal Size

Components, Phys. Chem. Chem. Phys., 1, 4329-4336.

Wertheim, M. S. (1984a). Fluids with Highly Directional Attractive Forces. I

Statistical Thermodynamics, J. Stat. Phys., 35, 19-34.

48
Wertheim, M. S. (1984b). Fluids with Highly Directional Attractive Forces.II.

Thermodynamics Perturbation Theory and Integral Equations, J. Stat. Phys., 35,

34-47.

Wertheim, M. S. (1986a). Fluids with Highly Directional Attractive Forces. III

Multiple Attraction Sites, J. Stat. Phys., 42, 459-476.

Wertheim, M. S. (1986b). Fluids with Highly Directional Attractive Forces. IV

Equilibrium Polymerization, J. Stat. Phys., 42, 477-492.

Werthrim, M. S. (1987). Thermodynamic Perturbation Theory of Polymerization, J.

Chem. Phys., 87, 7323-7331.

Van Pelt, A. (1992). Critical Phenomena in Binary Mixtures. Classification of Phase

Equilibria with the Simplified-Perturbed-Hard-Chain Theory, PhD Thesis, Delft

Univ. of Technology.

Vilmalchand, P. and Donohue, M. D. (1985). Thermodynamics of Quadrupolar

Molecules: The Perturbed-Anisotropic -Chain Theory, Ind. Eng. Chem. Fundam.,

24, 246-257.

Vilmalchand, P., Celmins Ilga and Donohue, M. D. (1986). VLE Calculations for

Mixtures Containing Multipolar Compounds Using the Perturbed Anisotropic

Chain Theory, AIChE J. 32, 1735-1738.

49